The ductus arteriosus (DA) is a fetal artery that allows blood ejected from the right ventricle to bypass the pulmonary circulation in utero. At birth, functional closure of the DA is initiated within minutes by O2-induced vasoconstriction. Functional closure (vasoconstriction) stops right to left shunting of blood and promotes anatomical closure. Failure of these processes leads to persistent ductus arteriosus, a common form of congenital heart disease in premature infants. During the first 5 years of this grant we showed that the DA's O2- sensing pathway consists of a sensor (the mitochondrial electron transport chain), which produces a diffusible mediator (H2O2), that inhibits voltage-gated K+ channels, such as Kv1.5. At birth, O2-induced increases in mitochondrial H2O2 in DA smooth muscle cells (DASMC) promote constriction by several mechanisms: Kv channel inhibition, direct activation of O2-sensitive calcium channels and rho kinase activation. Moreover, preterm DASMC are relatively deficient in these mechanisms, explaining the prevalence of persistent DA in preterm DA. This renewal focuses on a discovery made during a search for splice variants of Kv1.5 in human DASMC. We found a novel K+ channel, Human Oxygen-Sensitive K+ channel (HOSK), that when heterologously expressed creates a current that is voltage-gated, displays K+ specificity (Rb>K>>Cs>Na), and is 4- aminopyridine sensitive. HOSK appears to contribute to the resting membrane potential in human DASMC. HOSK siRNA reduces the O2-sensitive current in human DASMC. HOSK cDNA corresponds to a 3.0 kb neuronal, expressed sequence tag (EST) and has an unusual coding mechanism. HOSK and collagen 12(I) have identical mRNA with the much smaller 21 kDa HOSK resulting from initiation of translation at an internal ribosomal entry site (IRES). In silico modeling suggests that HOSK may have four hydrophobic domains (HD), a unique K+ selectivity filter (GVL, rather than the typical GYG amino acid sequence) and a variant voltage sensor. Phylogenetic analysis suggests HOSK originated in amniotes. In this proposal, the relative importance of HOSK versus canonical O2-sensitive voltage-gated K+ channels, Kv1.5 and Kv2.1, will be compared in term human DA, and two models of impaired O2 constriction: preterm rabbit DA and ionically remodeled human DA. Aim 1: HOSK is a novel K+ channel, arising independent of the canonical K+ channel family. Aim 2: HOSK contributes to DA constriction and augmenting HOSK expression will enhance O2- constriction in preterm rabbit DA and ionically remodeled human DA. Aim 3: HIF-11 shifts translation of the collagen gene away from HOSK toward collagen 21(1). HOSK, hidden by its complex encoding mechanism and unique structure, may offer a new explanation for how O2 causes constriction and will shed light on the link between constriction and fibrous obliteration of the DA. Since HOSK may have arisen independent of the canonical K+ channel family, exploring its structure-function relationship will extend our fundamental knowledge K+ channels.